小波
降噪
计算机科学
噪音(视频)
小波包分解
小波变换
算法
平稳小波变换
卫星
人工智能
离散小波变换
作者
Shulin Xiao,Lintao Han,Jiabian An,Luyao Gao,Changhong Hu
出处
期刊:Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
日期:2020-09-17
卷期号:: 517-530
标识
DOI:10.1007/978-3-030-69072-4_42
摘要
The principle of wavelet threshold de-noising is described in detail. Daubechies wavelet function is selected to denoise the sinusoidal signal with noise. The relationship between the order of wavelet function, the number of decomposition layers, the number of signal samples and the signal-to-noise ratio of denoised signal is demonstrated. At the same time, the calculation methods of fixed threshold, unbiased risk estimation threshold, mixed threshold and mini threshold max threshold are summarized, and then the denoising effects of hard threshold, soft threshold and semi soft threshold functions are compared. By analyzing the distribution of wavelet decomposition coefficients and combining with the characteristics of other threshold functions, a new wavelet threshold function is designed and compared with other threshold functions.
科研通智能强力驱动
Strongly Powered by AbleSci AI